An examination of the influence of mutual CFOaudit firm tenure on audit quality

If regulators’ concerns that the relation-ship that develops over time between an audit firm and their client’s CFO impairs auditor judgment are justified, then we should observe a negat

Trang 1

An examination of the influence of mutual CFO/audit firm

tenure on audit quality

Jeff L Paynea,⇑, Russell Williamsonb

a

University of Kentucky, Gatton College of Business and Economics, Von Allmen School of Accountancy, 423K GBE, Lexington, KY 40506, United States b

University of Louisville, College of Business, RW 273, Louisville, KY 40292, United States

a r t i c l e i n f o

Article history:

Available online xxxx

a b s t r a c t

This study examines whether the extent of professional relationships between an audit firm and their client’s CFO influences audit quality If regulators’ concerns that the relation-ship that develops over time between an audit firm and their client’s CFO impairs auditor judgment are justified, then we should observe a negative relationship between the length

of audit firm’s tenure with their client’s CFO and audit quality The results suggest that mutual audit firm-CFO tenure is associated with lower audit quality measured by the mag-nitude of discretionary accruals, the reduced incidence of issuance of going-concern audit opinions for distressed companies, and an increased likelihood of the receipt of an Accounting and Auditing Enforcement Release (AAER) from the US Securities and Exchange Commission (SEC) These affects are concentrated in a subsample of firms with higher levels of corporate governance concerns These findings have implications for poli-cies related to audit firm rotation Specifically, the results suggest that regulators need to consider other relationships underlying audit firm tenure, such as the relationships that form between audit firm and client personnel, when evaluating audit firm rotation policies

‘‘Independence, both historically and philosophically, is the foundation of the public accounting profession and upon its maintenance depends the profession’s strength and its stature” (Carey, 1970; 182).”

1 Introduction

The purpose of this study is to explore whether the extent of personal / professional relationships that develop over time – measured by the consecutive number of years that an audit firm and their client’s CFO work together – affect audit quality This research is motivated by a response from the Center for Audit Quality (CAQ) to the Public Company Accounting Over-sight Board’s request for public comment on auditor independence and audit firm rotations stating:

‘‘We [CAQ] also note that there are many existing factors that already limit the tenure of the engagement team and com-pany management In addition to the natural turnover within the audit engagement team, due to staff attrition and pro-motion, current independence requirements require lead audit partners and engagement quality control review partners

to rotate every five years Certain other partners involved with a company’s audit must be rotated after seven years

https://doi.org/10.1016/j.jaccpubpol.2021.106825

⇑Corresponding author.

E-mail addresses: jeff.payne@uky.edu (J.L Payne), russell.williamson@louisville.edu (R Williamson).

Contents lists available atScienceDirect

J Account Public Policy

j o u r n a l h o m e p a g e : w w w e l s e v i e r c o m / l oc a t e / j a c c p u b p o l

Please cite this article as: J.L Payne and R Williamson, An examination of the influence of mutual CFO/audit firm tenure on audit quality, J Account Public Policy,https://doi.org/10.1016/j.jaccpubpol.2021.106825

Trang 2

Moreover, there is a similar natural turnover of public company CFOs As detailed in a Crist/Kolder Associates 2011 study on executive management volatility, from January 1, 1995 through July 31, 2011, the average CFO .tenure within the S&P 500 was 5.1 .years We believe that the factors described above already limit the length of the relationships between the engagement team and company management and stifle any opportunity for ’coziness’ between the auditor

This research empirically examines this statement to ascertain its validity and inform discussions regarding mandating audit firm rotation

We investigate mutual audit firm / CFO tenure as another setting to investigate potential independence impairment In a report released in August of 2011, the PCAOB expressed concern that inspection reports continue to indicate a failure of audi-tors to exercise appropriate levels of objectivity and professional skepticism during the audit of their clients’ financial state-ments (PCAOB 2011a) Although the PCAOB cannot document that this loss of objectivity is directly related to long audit firm tenure, questions nevertheless persist (PCAOB 2011a) Academic investigation likewise produces inconsistent results regard-ing the effects of long audit firm tenure on professional skepticism and ultimately on audit quality Although a few recent studies report decreased audit quality, most research finds audit quality improves as audit firm tenure increases The PCAOB and the accounting profession can be informed by additional investigation of this issue

An important aspect of the auditor-client relationship is the professional interaction between the auditing firm and their client’s personnel, specifically the CFO CFOs are directly responsible for the company’s financial statements (Gibbins et al

2007), negotiate with the audit partner regarding the fair presentation of reported balances (Gibbins et al 2005), and nor-mally have compensation and equity incentives based on operating results Because of their role in financial reporting, CFOs and auditors interact frequently Further, the CFO has more direct communication with the audit committee than any other party, and the CFO participates in the setting of the audit committee agenda 75% of the time This gives the CFO the ability to broker knowledge between the auditor and the audit committee, and the ability to engage in negotiations with the auditor before information reaches the audit committee

The Metcalf Report notes, ‘‘long association between a corporation and an accounting firm may lead to such a close iden-tification of the accounting firm with the interests of its client’s management that truly independent action by the account-ing firm becomes difficult.” (U.S Congress 1976, 19) In addition, out of concern of relationships forming between audit firm personnel and client’s personnel the Cohen commission stated, ‘‘many of the asserted advantages of rotation can be achieved

if the public accounting firm systematically rotates the personnel assigned to the engagement.” (AICPA 1978, xxx) Further,

at a meeting of the PCAOB’s Investor Advisory Group (IAG), some members of the IAG advocated mandatory firm rotation stating that ‘‘key to concern over independence was the level of ‘coziness’ the firm had with the management of the company being audited.” (PCAOB 2011b, 6)

On the other hand, the mutual trust that develops over time can improve inter-organizational communication, and in turn benefit the financial reporting process If the negative aspects associated with long audit firm tenure dominate and are due to the ‘‘coziness” with client management then a frequent break-up in this relationship (either due to CFO turnover

or audit firm rotation) should strengthen auditor independence and skepticism The average mutual tenure between and audit firm and their client’s CFO tenure for U.S public companies of approximately three years likely creates a setting where sufficient time is not available for strong relationships and potential biases or trust to form The ‘‘break” in the relationship between audit firm and CFO may reduce the influence that long audit firm tenure may have on audit quality In the end, it could be that such breaks, if frequent enough, can provide a reasonable alternative to mandatory audit firm rotation Prior research investigating the influence of audit firm tenure on audit quality has not considered that underlying the audit firm’s tenure is a subset of important relationships that form and break between client and audit firm personnel

We single out one set among many relationships: the tenure of professional relationships between a client CFO and their audit firm By analyzing this relationship and its effect on audit quality our investigation extends the understanding of the influence of audit firm tenure on audit quality by identifying the influence of these relationships

Specifically, we investigate audit quality measured as 1) the absolute value of discretionary accruals, and separately the income increasing and income decreasing discretionary accruals present in reported financial statements, 2) the issuance of

a going-concern opinion for distressed companies, and 3) the receipt of an Accounting and Auditing Enforcement Release (AAER) from the US Securities and Exchange Commission (SEC) For each examination, we use the complete time period where data is available

The results indicate that a lengthening of the mutual tenure between an audit firm and their client’s CFO leads to higher levels of discretionary accruals, measured using absolute total and income increasing accruals.1There is also a decrease in the issuance of going concern opinions for distressed companies and an increase in AAERs for clients with elevated levels of mutual tenure Additionally, when we split our samples between high and low quality corporate governance firms we find that, gen-erally, mutual tenure is associated with reduced audit quality in firms with higher numbers of corporate governance concerns than strengths

We provide a additional tests investigating the influence of mutual tenure on the presence of material weaknesses in the internal control over financial reporting (ICFR) report and additional untabulated analyses examining the influence of short

1

We analyze both absolute abnormal accruals and subsamples of income-increasing and income-decreasing abnormal accruals because Hribar and Nichols (2007) demonstrate that the analysis of absolute accruals may be problematic due to a correlated omitted variable problem.

2 Please cite this article as: J.L Payne and R Williamson, An examination of the influence of mutual CFO/audit firm tenure on audit quality, J Account Public Policy,https://doi.org/10.1016/j.jaccpubpol.2021.106825

Trang 3

periods of mutual tenure, the presence of financial restatements from fraud, and differential affects for companies in high risk industries The results of these tests generally support the implications of our main results, that mutual tenure nega-tively influences audit quality and suggest that a significant portion of audit quality concerns arising from auditor tenure are a function of the mutual tenure between the CFO and the audit firm and that this effect is concentrated withing firms with lower quality corporate governance

This paper makes an important contribution to the literature By examining the relationships that develop between audit firms and their client’s CFOs, we contribute new insights to the debate on audit firm rotation We contribute to the literature

by examining the bonding relationship between auditor and CFO We suggest this is an important modification to prior research that more directly examines the role of relevant auditor tenure The PCAOB is concerned that audit firms that iden-tify with their clients because of a long association can compromise audit quality The solution prescribed is to require mandatory audit firm rotation to ‘‘break” this relationship However, the extant research suggests that perhaps it is not the long audit firm tenure per se that might lead to lower audit quality but the underlying relationships between audit firm and client personnel that can give rise to professional or even personal ties which in turn can contribute to lower auditor independence We examine such relationships and find that increasing audit firm tenure improves audit quality, but the length of the relationship between audit firm and the client CFO weakens it Perhaps this suggests a more sensible approach

to deriving the perceived benefits from mandatory audit firm rotation, targeting those cases where the audit firm has formed and maintains long-term ties to client personnel Thus, an audit firm’s tenure with their client’s CFO is an important relation-ship to consider when evaluating audit firm tenure effects on audit quality

The remainder of the paper is organized as follows.Section 2describes the background research and states the hypoth-esis.Section 3describes the method, analyses, discusses limitations, provides supplemental analysis, and Section 4 con-cludes the paper

2 The debate surrounding audit firm rotation and empirical evidence

2.1 Audit firm rotation

Commission, 2010, 2011, 2013)2and past (AICPA 1978, GAO 2003) debate Recently, the US House of Representatives passed

a bipartisan bill, H R 1564 that prohibits the Public Company Accounting Oversight Board (PCAOB) from requiring mandatory audit firm rotation (US Congress 2013).3Proponents of audit firm rotation argue it provides a ‘‘fresh look” at the client’s control environment and financial reporting decisions and lessens both the economic and personal relationships between the audit firm and the client (PCAOB 2011a) The PCAOB is motivated by its own inspections of audit firms that indicate a failure of audit firms

to exercise appropriate levels of objectivity and skepticism However, the PCAOB has not been able to establish that loss of objectivity and skepticism is due to a long audit firm tenure (PCAOB 2011a)

Those that oppose mandatory audit firm rotation argue that it sacrifices the formation of knowledge and expertise (as discussed in PCAOB Auditing Standard #5) that auditors develop during the performance of consecutive audits and that it can increase audit production cost (US Congress 2013) From the PCAOB call for comments on mandatory audit firm rotation, 95% of respondents opposed its implementation primarily on the grounds that rotation would negatively affect both quality and efficiency of an audit (Cohn 2012; Hanson 2013)

Extant research on this debate provides some evidence that does not support the mandatory rotation argument; however,

mandatory rotation.4Reid and Carcello (2017)find a negative market reaction to events that potentially increased the likeli-hood of regulations requiring mandatory rotation, especially for companies with longer audit firm tenure.Fiolleau et al (2013)note some additional potential consequences to mandatory audit firm rotation First, it would provide management with additional opportunities during auditor change years to seek a desired financial reporting outcome from other bidding audit firms Therefore, required rotation might actually escalate management’s ability to influence auditors Second, required rotation allows management to change auditors without having to disclose disagreements or problems If disclosures related to auditor rotation are useful to financial statement users (Hennes et al 2014), required rotation would remove this information for audi-tor changes that occur in accordance with the regulation Opponents recommend that the PCAOB not require mandaaudi-tory audit

2

In the revised version of the 8th Directive on Statutory Audits of Single Annual and Consolidated Financial Statements ( European Union 2006 ), the European Union (EU) requires that by the end of June 2008, all twenty-seven-member states of the European Union enact the requirements of the revised 8th Directive into national law One important detail of the Directive is rotation of the key audit partner Every member state is given discretion regarding the length of the rotation period.

3

The bill passed with a vote of 321 for, 62 against The bill did not move forward in the US Senate.

4

They estimate[d] consumer surplus losses at approximately $2.4–3.6 billion (total audit fees in 2010 were $11 billion) if rotation were required after ten years, $4.3–5.5 billion for a four-year rotation policy.

Trang 4

firm rotation until evidence that points to a relation between audit firm tenure and inspection report findings is documented (Anton and Melancon 2011)5and the cost/benefit relationship is better understood (GAO 2003).6

An outcome of this continued discussion was addressed in 2017 when the PCAOB issued AS 3101: The Auditor’s Report on

an Audit of Financial Statements When the Auditor Expresses an Unqualified Opinion (PCAOB 2017) A part of this new stan-dard requires auditors to indicate the first year they were engaged by the audited company in the audit report While not requiring audit firm rotation, this does provide audit tenure information to financial statement users This portion of the standard was not adopted without controversy Investors suggested this information would be useful in the ratification of audit firms, but others suggested that existing research did not find a consistent relationship between audit tenure and audit quality and were concerned that the difficulty at determining the start date of some clients, for example due to audit firm mergers, would reduce the value of the information (SEC 2017) Board member Jeanette Franzel voted against the provision based on concerns that investors would infer an audit quality/tenure connection that might not exist (Hirschmann 2017) In making the final decision the Board concluded that the provision of the information is an efficient way to inform retail inves-tors without their need to search elsewhere, often at a cost, for this information

2.2 Audit firm tenure and audit quality

Existing research examines the association between audit firm tenure and various proxies for audit/financial reporting quality including discretionary or unexpected accruals, reporting earnings to meet or beat earnings targets, the cost of debt, investor response to reported earnings, financial statement restatements, and others.Johnson et al (2002), andGul et al (2009)find that accrual quality is unaffected by audit firm tenure.Stanley and DeZoort (2007)find a negative relation between the length of the auditor–client relationship and the likelihood of restatement.Myers et al (2003) and Blouin

et al (2007)find that accrual quality increases in audit firm tenure.Li (2010)documents a positive association between the conservatism in reported earnings and the length of the auditor–client relationship In their investigation of audit partner and audit firm tenure using international data,Chen et al (2008)also do not find a negative effect of tenure on earnings qual-ity Using data from internal audit firm inspections,Bell et al (2015)document a positive relationship between tenure and audit quality among SEC clients.Davis et al (2009)examine audit quality by investigating how audit firm tenure influences management’s ability to use discretionary accruals to report earnings that meet or beat analysts’ earnings forecast They find longer audit firm tenure reduces the use of discretionary accruals to meet or beat analysts forecast, an indication of improved audit quality However, they find a turning point for audit tenure of greater than 15 years, where tenure leads to a reduction

of audit quality The non-linear effect of audit firm tenure on audit quality is confirmed byBrooks et al (2016).Singer and Zhang (2018)find that SOX has mitigated, but not eliminated the negative effects of long-term tenure that they find leads to less timely discovery and adjustment of misstatements They also find the magnitude of the misstatements are positively correlated with increasing audit firm tenure.Yen et al (2018)investigate a unique setting to the literature and find that audit firms with more experience with a client is better able to assess information security breech risks.Chu et al (2018)find that longer audit firm tenure for clients with increased litigation risk reduces audit quality Investigating short tenure relation-ships, results using various proxies for audit quality consistently indicate that quality is diminished during the first years of

an audit engagement (AICPA 1992; Geiger and Raghunandan 2002; Carcello and Nagy 2004; Johnson et al 2002; Davis et al 2009; Bell et al 2015)

A goal of mandatory audit firm rotation is to reduce the length of the relationship between the audit firm and client lead-ing to increased professional skepticism and improved audit quality The inconsistent results from existlead-ing literature leave the profession and regulators with little information to support requiring mandatory audit firm rotation However, some research does find settings where long audit firm tenure reduces audit quality Importantly, the consistent finding that audit quality is diminished in the initial years of an audit/client relationship supports opponents concerns that mandatory audit firm rotation leads to reduced audit quality Importantly,Gipper et al (2018)find little support for the potential ‘‘fresh look” benefits of mandatory five-year partner rotations in a new study using proprietary PCAOB data from audit firm inspections 2.3 Chief financial officers

Extant literature investigating the influence of audit firm tenure on audit quality overlooks a critical participant in the relationship between the audit firm and their client, the client’s chief financial officer (CFO).7CFO’s typically have primary

5 Since PCAOB Inspection reports do not reveal the clients where deficiencies occur, this analysis cannot be performed with publicly available data 6

KPMG notes that countries such a Singapore, South Korea, Argentina, Brazil, Spain and Canada have implemented mandatory audit firm rotation (MAFR) and subsequently either partially or fully withdrew it.Additionally, countries such as Australia, Hong Kong, Japan, Malaysia, Mexico, New Zealand, Russia, Sri Lanka, Switzerland, Thailand and the USA have considered MAFR and decided against the adoption of MAFR ( KPMG 2017 )

7

Fiolleau et al (2013) find that audit firms actively align attributes of their key personnel with the client’s CFO early in the auditing bidding process.

Trang 5

responsibility for decisions that influence their firm’s financial reporting process and related financial reports (Ge et al 2011; Menon and Williams, 2008).8 , 9CFOs participate in auditor–client negotiations related to accounting disputes (Geiger and North 2006; Gibbins et al 2005) These negotiations take place between the CFO and the auditors throughout the audit process and before the financial statements and audit results are communicated to the audit committee Likewise, the CFO has incentives to maximize their compensation by not disappointing investors with unsatisfactory financial reports (e.g., missing analysts fore-cast) (Jiang et al 2010)10 and face a greater likelihood of dismissal when reportable events are made public (Menon and Williams 2008) or company performance decreases (Coughlan and Schmidt, 1985; Gilson, 1989; Beneish, 1999; Mian 2001, Burks 2010)

The Independence Standards Board (ISB 2000) report expresses concerns that the relationship between the audit firm and client personnel can lead to biased decisions, with the auditor potentially accepting the preferences of the client over the preferences of the audit firm And since the CFO’s influence on the financial reporting process puts them in frequent ongoing contact with the auditor (Antle and Nalebuff, 1991) and the CFO plays a key role in audit committee communication and organization (Beasley et al 2009), the CFO-auditor relationship, consequently, has the greatest potential to affect audit qual-ity among the various levels of management inside the audit client

Because of the important roles the CFO and audit firm play in the financial reporting process, we focus primarily on the audit firm - CFO relationship and how it influences audit quality Given the agency and the mutual trust perspectives on the mutual role of audit firm and CFO, it is not immediately clear as to how a long association between the audit firm and CFO will influence financial reporting and audit decisions On the one hand, from an agency perspective, CFOs have incentives to manage earnings and exert some form of pressure (economic or social) on auditors, potentially resulting in biased reporting and low audit quality

On the other hand, the mutual trust perspective suggests the social interactions that develop with time can enhance the infor-mation flows between the two parties in turn improving the quality of auditing (Dees and Cramton, 1991)

In summary, prior literature provides mixed evidence on the effect of audit firm tenure on audit quality These inconsis-tent findings suggest the need to carefully examine the underlying relationships between audit firm and their client’s per-sonnel to identify conditions where long audit firm tenure might be problematic Based on the concerns documented in the extant literature and by the regulatory agencies regarding the detrimental aspect of these personal relationships we hypothesize:

Hypothesis: There is a negative relationship between mutual audit firm – client CFO tenure and audit quality

2.4 Mutual tenure and corporate governance

Where prior research on auditor tenure has found auditor tenure to be associated with measures of reporting quality such

as accounting conservatism (Jenkins and Velury, 2008) a deeper look into this relationship shows that the association with higher quality information is limited to sub-sets of the population and that negative associations exist for firms with fewer resources or firms with weaker external monitoring (Li, 2010) These concerns regarding long auditor tenure are often pre-sented as independence concerns, however,Stanley and DeZoort (2007)find that non-audit fees have no significant relation-ship with the likelihood of restatement for long tenure auditors, suggestions independence may not be impaired Further,

Kaplan and Mauldin (2008)find within an experimental setting that independence judgements center more upon the quality

of the firm’s governance mechanisms than on the auditors tenure.11Since, auditor selection and rotation is a function of cor-porate governance and corcor-porate governance has been shown to affect the quality, transparency, and frequency of corcor-porate disclosures (Eng and Mak, 2003; Wang and Hussainey, 2013; Haniffa and Cooke, 2005; Kelton and Yang, 2008), we ask the open research question:

Research Question: Will mutual tenure between the Auditor and the CFO have differential effects for high quality corporate governance firms as opposed to low quality corporate governance firms?

3 Models and empirical results

3.1 Descriptive statistics

All model variables are defined inTable 1 Panel A presents the variable definitions, Panel B outlines our sample construc-tion for each test, Panel C provides the descriptive statistics and Panel D the Pearson correlaconstruc-tions In our sample, the average audit firm tenure is 10.58 years, the average CFO tenure is 3.80 years, and the average mutual tenure is 3.34 years

8

CFO’s have unique characteristics when compared to CEOs ( Crist/Kolder Associates 2017 ) CFO’s employment relationship turns over more frequently and very few CFOs (6%) become future CEOs Interestingly, only 25% of sitting CFOs at the time of their study had accounting degrees and 32% of these have work experience in a larger international accounting firm (commonly denoted the Big-N).

9

Krishnan and Wang (2015) examine the relation between managerial ability, i.e., ability in transforming corporate resources to revenues, and audit fees and

a going concern opinion and find that incremental to firm-level attributes, both audit fees and the likelihood of issuing a going concern opinion are decreasing

in managerial ability.

10

They report ‘‘ that the magnitude of accruals and the likelihood of beating analyst forecasts are more sensitive to CFO equity incentives than to those of the CEO” (2010, p 513).

11

Specifically that non-professional investors believe that auditors will be more independent under a strong audit committee regardless of auditor tenure.

Trang 6

Table 1

Panel A: Variable Definitions.

Variables Definition

Dependent Variables

ABSDA Following Dechow, Sloan, and Sweeney (1995) , and Kothari, Leone, and Wasley (2005) , the estimates of k 1 ; k 2 and k 3 are those obtained from the original Jones model where total accrual is

earnings before extraordinary items and discontinued operations minus the operating cash flows (IB-OANCF) DSALE t is the change of total revenue from t 1 to t year, DREC t is the change of net receivables from t 1 to t year, and PPE is the property, plant and equipment, ROA is the net income (NI) divided by total assets (AT) (Compustat)

Total Accrualt

ATt1 ¼ k 1 ð 1=AT t1 Þ þ k 2 ðDSALE t DREC t Þ þ k 3 ð PPE t Þþk 4 ð ROA t Þ þet

GC Equal to one if the firm received a going concern opinion in the current year, 0 otherwise (Audit Analytics)

AAER Equal to one if the firm was included in an Accounting and Auditing Enforcement Release (AAER) from the US Securities and Exchange Commission (SEC).

Main Independent Variables

AUDTEN = the number of years the auditor has been retained by the client beginning in 1980 (Compustat)

CFOTEN = the number of years the client’s employee has been the chief financial officer (Execucomp and Audit Analytics)

MUTUAL = the number of years with the same audit firm and client’s chief financial officer.

Control Variables

AGE = Natural logarithm of the number of years that the client firm has been listed on COMPUSTAT Source: Compustat

ASSETS =the natural log of total assets (AT), (Compustat)

BIGN =a dummy variable that equals one if the company uses a big-N audit firm; zero otherwise 19

(Compustat) CASHFLOW = operating cash flow (OANCF) scaled by total assets (AT), (Compustat)

FIN =a dummy variable indicating mergers or new financing and equals one if COMPUSTAT footnote SALE_FN equals ‘‘AB”, or the percentage change in long-term debt (DLTT) is greater or equal to

20 percent, or the percentage change in common shares outstanding (CSHO), adjusted for stock splits, is greater or equal to 10 percent; zero otherwise, LCA =the absolute value of lagged current accruals, and

LEV = total liabilities (AT – CEQ) scaled by lagged total assets, (Compustat)

LITIG = a dummy variable equal to one if the company-year is in a high litigation industry, defined as SIC codes: 2833–2836, 3570–3577, 3600–3674, 522–5961, 7370–7474; zero otherwise,

(Compustat) LOSS = a dummy variable that equals one if net income (NI) is less than zero; zero otherwise (Compustat)

MB = market-to-book ratio (MKVALT/CEQ), (Compustat)

D2D = the distance-to-default metric is a default risk score based on functional for of the Merton model and described in Bharath and Shumway (2008)

MV = the natural log of the market value of equity (MKVALT) at fiscal year-end, (Compustat)

NEG EQUITY

=

an indicator variable equals 1 if the client firm has a negative book value of equity (Compustat) ROA =net income before extraordinary items scaled by lagged assets (Compustat)

SOX = a dummy variable that equals one for years after 2004.

Panel B: Samples Construction

To construct our analysis samples, we start with observations from non-financial services industries listed in Compustat from 1994 to 2015: 175,227 Next, we utilize Execucomp to calculate the tenure of the firm CFO and utilize officer information from Audit Analytics to supplement our sample when firms are missing from Execucomp Joining

this data set to our main Compustat sample reduces our sample by:

(95,495) Next, we calculate our variables; reducing our sample for firm year observations with missing data for the calculation of control variables: (42,600)

We then join Going Concern data from Audit Analytics for our dependent variable of interest and both daily and monthly returns data from CRSP to calculate additional control variables; reducing

our sample by:

(9464)

We then retain only firms with negative cashflow or negative earnings as distressed firms; reducing our sample by: (21,165)

Next, to maintain sample consistency we utilize our ABDSA sample and join data on Accounting and Auditing Enforcement Releases (AAERs) from the SEC and data for additional related control

variables (ASSETS, ROA, MERGER); reducing our sample by:

(5556)

Trang 7

Panel C: Descriptive statistics

Panel D Pearson’s pairwise correlation coefficients for the full sample of observations

ABSDA 1 1.00

AUDTEN 2 0.11 1.00

CFOTEN 3 0.08 0.22 1.00

MUTUAL 4 0.08 0.38 0.88 1.00

CASHFLOW 5 0.36 0.12 0.09 0.10 1.00

LEV 6 0.22 0.01 0.04 0.05 0.41 1.00

LITIG 7 0.07 0.07 0.05 0.04 0.06 0.06 1.00

MB 8 0.02 0.02 0.01 0.00 0.06 0.13 0.07 1.00

MV 9 0.30 0.30 0.17 0.20 0.40 0.22 0.04 0.16 1.00

LOSS 10 0.28 0.16 0.13 0.14 0.46 0.19 0.10 0.06 0.48 1.00

FIN 11 0.08 0.04 0.05 0.05 0.12 0.08 0.03 0.01 0.05 0.10 1.00

LCA 12 0.25 0.08 0.07 0.09 0.43 0.36 0.03 0.03 0.23 0.18 0.05 1.00

BIGN 13 0.28 0.28 0.08 0.12 0.32 0.16 0.03 0.04 0.58 0.29 0.06 0.23 1.00

SOX 14 0.16 0.05 0.12 0.12 0.14 0.06 0.04 0.02 0.15 0.14 0.03 0.07 0.28 1.00

AAER 15 0.01 0.01 0.02 0.02 0.01 0.01 0.02 0.02 0.04 0.01 0.02 0.01 0.03 0.07 1.00

ASSETS 16 0.37 0.30 0.17 0.20 0.46 0.25 0.14 0.03 0.88 0.42 0.03 0.32 0.60 0.15 0.03 1.00

ROA 17 0.05 0.00 0.01 0.01 0.12 0.27 0.01 0.03 0.06 0.04 0.02 0.12 0.05 0.02 0.00 0.12 1.00

NEG_EQUITY 18 0.20 0.05 0.06 0.06 0.27 0.51 0.00 0.32 0.25 0.23 0.04 0.20 0.15 0.07 0.02 0.23 0.08 1.00

GC 19 0.31 0.11 0.10 0.11 0.47 0.45 0.01 0.08 0.43 0.33 0.07 0.34 0.32 0.08 0.02 0.44 0.10 0.40 1.00

D2D 20 0.02 0.08 0.06 0.06 0.11 0.36 0.11 0.12 0.26 0.27 0.03 0.04 0.05 0.01 0.01 0.02 0.15 0.19 0.16 1.00 AGE 21 0.19 0.44 0.22 0.22 0.14 0.00 0.21 0.02 0.38 0.23 0.06 0.13 0.20 0.02 0.02 0.45 0.02 0.06 0.15 0.02 Dummy variables are included to control for year and industry fixed effects.

Coefficients in bold and italics indicate significance at the five percent level All variables are formally defined in Appendix B.

19 Big-N firms are Arthur Andersen, Deloitte, Ernst & Young, KPMG, and PricewaterhouseCoopers.

Trang 8

3.2 Method and analysis

To examine the hypothesis, we followDeFond and Zhang’s (2014)suggestion to examine multiple proxies for audit qual-ity They encourage the examination of measures on the opposite end of the ‘‘egregiousness” spectrum to better ascertain their impact on audit quality and suggest there is a gradient of causality to these proxies for audit quality Therefore, we examine three measures of audit quality substantiated in prior literature: 1) discretionary accruals, 2) going concern issu-ance for distressed companies, and 3) the receipt of an AAER Examining abnormal or discretionary accruals provides insights into management’s discretion in the application of GAAP to their reported financial statements.Lawrence, Minutti-Meza, and Zhang (2011)suggest this measure can reveal auditors influence on these decisions Discretionary accruals are powerful tools to alter reported earnings as they can be modified after year-end and right up to the earnings announcement date (Bratten et al., 2014) They do not require a change in business operations or strategy; they are simply changes to the finan-cial statement presentation of those operations

We also examine the overall likelihood of issuing a going concern opinion as a measure of quality and independence Prior research of both auditor independence and audit quality utilize the likelihood of issuing a going concern opinion as an oper-ational measure (Carson et al 2013) Specifically, the willingness to issue a going concern opinion is considered a significant measure of auditor independence (Chen et al., 2013), and a measure of audit quality (DeFond and Zhang 2014) Auditors risk reputational damage and litigation costs when failing to issue a going concern, and therefore, audit firms will resist manage-ment’s desire to avoid the going concern statement and report in a more conservative fashion (DeFond and Francis, 2005)

We model the overall likelihood of issuing a going concern opinion to investigate the association between mutual auditor CFO tenure on audit quality and independence

Additionally, this study employs AAERs that provide a potentially more relevant and direct measure of audit quality by measuring actual outcomes of the audit process (DeFond and Zhang 2014; DeFond and Francis 2005) AAERs report civil lit-igation and administrative proceedings by the SEC’s and represent severe violations of Generally Accepted Accounting Prin-ciples (Lennox and Pittman 2010)

3.3 Discretionary accruals

For our first measure of audit quality we use discretionary accruals These represent abnormal accruals and could indicate attempts by management to alter reported financial statement balances The sample for this analysis starts with all obser-vations available in non-financial service industries for the years 1994–2015 from Compustat (175,227).12We next collect CFO data from Execucomp and Audit Analytics Since CFO’s are not always tracked on Execucomp (Brochet et al., 2011), we select unique additional observations provided in the Audit Analytics D&O changes database Joining this data set with our ini-tial sample eliminates 95,495 observations After eliminating observations without sufficient data to calculate discretionary accruals (42,600) final sample contains 37,132 observations (seeTable 1, Panel B)

FollowingDechow et al (1995), andKothari et al (2005), the estimates of k1; k2 and k3are those obtained from the orig-inal Jones model where total accruals are earnings before extraordinary items and discontinued operations minus the oper-ating cash flows (IB-OANCF).DSALEtis the change of total revenue from t 1 to t year,DRECtis the change of net receivables from t 1 to t year, and PPE is the property, plant and equipment ROA is the net income (NI) divided by total assets (AT)

Total Accrualt

ATt1 ¼ k1ð1=ATt 1Þ þ k2ðDSALEtDRECtÞ þ k3ðPPEtÞþk4ðROAtÞ þet

Consistent with prior studies, we, winsorize all variables at the one percent tails before estimating equation (1) within years and 2-digit SIC codes (excluding industries with less than six members)

Discretionary accruals, DA is equal to the residual values from estimating equation (1) Absolute discretionary accruals, ABSDA, is equal to the absolute value of DA Consistent with prior studies, we eliminate observations with ABSDA greater than one We estimate equation (2) to test for a relationship between audit quality and audit firm tenure (AUDTEN), CFO tenure (CFOTEN), and mutual tenure (MUTUAL) to capture the time period the CFO and audit firm worked together

ABSDAt¼u0þu1AUDTENtþu2CFOTENtþu3MUTUALtþu4CASHFLOWtþu5LEVtþu6LITIGtþu7MBt

þu8MVtþu9LOSStþu10FINtþu11LCAt 1þu12BIGNtþu13SOXtþu14YearFEtþu15IndustryFEtþe ð2Þ

ABSDA as defined earlier, is the proxy for audit quality AUDTEN captures the number of years the auditor has been retained by the client beginning in 1980 CFOTEN captures the number of years a CFO worked with a company MUTUAL

is the variable of interest and measures the mutual effect of audit firm tenure and CFO tenure on audit quality If mutual audit firm-CFO tenure leads to reduced audit quality, then the coefficient on MUTUAL will be positive We also include con-trols firm operating cashflow, leverage, and negative earnings to control for firm financial performance (CASHFLOW, LEV, LOSS), we include an indicator for firms in high litigation risk industries (LITIG) and a the log transformed market value of the firm (MV) to control for firm complexity and inherent risk Further, we include both the market-to-book value of the firm (MB) to control for firm growth and its effect on accruals and (FIN) to control for mergers or new financing In addition to cash

12

The sample period in 1994 as this is the first year that 100 or more observations with available CFO data is available.

Trang 9

flows, we also control for the prior year’s absolute value of current accruals (LCA) both of which directly affect the cashflow-accrual relationship We include controls for characteristics of the audit environment, specifically if the firm is a Big N audi-tor (BIGN) and if the firm year observation falls after the implementation of Sarbanes-Oxley (SOX) Finally, this model includes both industry fixed effects at the two-digit SIC code level and year fixed effects as well as firm clustered standard errors to control for firm, industry, and time specific variations in accruals

4 Results

4.1 Discretionary accruals

We present the descriptive statistics of the discretionary accruals sample inTable 1, Panel C The mean (median) value of ABSDA is 0.10 (0.06) and is consistent with levels reported in prior research The mean (median) values of AUDTEN and CFO-TEN are 10.58 (9.00) and 3.80 (3.0) years, respectively Panel C provides a correlation matrix that indicates auditor tenure and CFO tenure are negatively associated with the absolute value of discretionary accruals.13

Table 2, Panel A presents the regression results of the discretionary accruals model (equation 2) using the absolute value

of total accruals (ABSDA) The OLS results for our investigation of discretionary accruals, presented with our mutual tenure term of interest in Column B, and without in column A have adjusted R-square of 0.1966 and 0.1967 respectively These val-ues suggest a reasonably good fit and are comparable to levels reported in prior research

Panel A Column A reports the results of our discretionary accruals model without the term employed to capture the effect

of mutual auditor/CFO tenure This model reports that both auditor tenure and CFO tenure are significantly negatively asso-ciated with discretionary accrual levels (AUDTEN,0.0003, p < 0.01; CFOTEN, 0.001, p < 0.01) The coefficient on AUDTEN is negative and significant, indicating higher audit quality with longer audit firm tenure consistent some prior research (e.g.,

Myers et al 2003) The negative and significant coefficient on CFOTEN suggests that the CFOs’ increased tenure is associated with lower levels of accruals-based earnings management Panel A Column B reports the results of our model including the MUTUAL capture the effect of mutual auditor/CFO tenure associated with levels of discretionary accruals We find that, hold-ing the levels of CFO and auditor tenure constant, the effect of another year of mutual auditor/CFO tenure is positive and significant (MUTUAL, 0.001, p < 0.05), suggesting that increasing mutual tenure between the auditor and CFO reduces audit quality as measured by increased levels of discretionary accruals

4.2 Discretionary accruals, mutual tenure, and corporate governance

To explore our research question regarding corporate governance, we reexamine our mutual tenure and discretionary accrual model using cross sectional variation in our sample related to the quality of the firm’s corporate governance to iden-tify high quality and low quality governance firms We then investigate if mutual tenure affects our measures of audit quality differently depending on the quality of the firm’s corporate governance To measure corporate governance we utilize the MSCI corporate governance index; MSCI generates a variety of responsibility metrics for publicly traded firms in order to build their socially responsible investment portfolios This database has been used extensively to examine corporate envi-ronmental and social concerns (Asante-Appiah, 2020; Ballou et al., 2018, Hummel and Schlick, 2016) and corporate gover-nance issues (Kim et al., 2012; Cho et al., 2013)

The MSCI Corporate Governance index includes measures of both governance strengths and concerns including audit committee independence, board of directors’ attendance, presence of poison pill provisions, etc.14Using the MSCI corporate governance index we split our sample into ‘concern’ and ‘non-concern’ firms where concern firms are any firm where the MCSI reports a greater number of weaknesses than strengths.15

InTable 2Column C and D we report the results of our corporate governance (CG) tests within our CG concern and CG non-concern subsamples, respectively We find in column C that the association between discretionary accruals and mutual auditor/CFO tenure is positive and significant (Col C, MUTUAL, 0.002, p < 0.05) within our subsample of ‘concern’ firms and

we do not find a significant association reported in column D within our ‘non-concern’ sample This association suggests that within firms having higher quality corporate governance, mutual tenure does not contribute to an increase in absolute dis-cretionary accruals while lower quality corporate governance is associated with increasing disdis-cretionary accruals

In addition to our corporate governance sub-sample tests, we also split our sample between signed positive and negative

income-increasing accruals (POSDA) the coefficient on MUTUAL is positive and significant (0.001, p < 0.10), however, for income-decreasing accruals, the coefficient is not significant (0.001, p > 0.10) To the extent that income increasing

discre-13

The remaining control variables have values consistent with prior research and the correlations between the remaining independent variables are well within acceptable levels and do not indicate the presence of multicollinearity.

14

The methodology to generate the MSCI Corporate Governance Index may be found here: https://www.msci.com/eqb/methodology/methdocs/MSCI_ Governance-Quality_Jun15.pdf

15

The ‘non-concern’ firms in this sample include ‘balanced’ firms where net strengths-weaknesses = 0; we find consistent results if we choose to include

‘balanced firms’ with concern firms or omit them from the model.

Trang 10

tionary accruals represent opportunistic reporting, our results provide some evidence that mutual tenure is associated with this behavior

The results from these analyses indicate that as the mutual tenure of the CFO and audit firm increases, audit quality is reduced as more discretionary accruals are used by management in their reported financial statements In other words, financial statement quality, as measured by management’s use of discretionary accruals, is lessened as mutual tenure increases Further, our subsample investigation suggests that corporate governance has an important and significant role

to play in determining how mutual tenure affects audit quality

Table 2

Audit Firm Tenure, CFO Tenure and Discretionary Accruals (ABSDA) Panel A.We estimate equation (2) to test for a relationship between audit quality and audit firm tenure (AUDTEN), CFO tenure (CFOTEN), and mutual tenure (MUTUAL) to capture the time period the CFO and audit firm worked together.ABSDA t

=-t u 0 + u 1 AUDTEN + u 2 CFOTEN + u 3 MUTUAL + u 4 CASHFLOW + u 5 LEV + u 6 LITIG + u 7 MB + u 8 MV + u 9 LOSS + u 10 FIN + u 11 LCA + u 12

***, ** and * denote statistical significance at the 1, 5 and 10 percent levels, respectively, using two-tailed tests The t-statistics reported in parentheses are based on standard errors that are heteroskedasticity robust and clustered at the firm level Regressions include year and industry fixed effects Column A reports the results of our model without our variable of interest, MUTUAL, which is included in our model reported in column B Columns C and D report regression results when the MCSI governance score is negative (Col C) or non-negative (Col D) respectively.

Panel B: Audit Firm Tenure, CFO Tenure and Discretionary Accruals (POSDA and NEGDA)

***, ** and * denote statistical significance at the 1, 5 and 10 percent levels, respectively, using two-tailed tests The t-statistics reported in parentheses are based on standard errors that are heteroskedasticity robust and clustered at the firm level Regressions include year and industry firm fixed effects This table uses the ABSDA sample with 37,132 observations, to examine how audit firm tenure (AUDTEN), CFO tenure (CFOTEN) and the period of mutual tenure (MUTUAL) influence reported accrual balances ABSDA is calculated following Dechow, Sloan, and Sweeney (1995) , and Kothari, Leone, and Wasley (2005) Variable definitions are in Table 1 , Panel A.

Định dạng
Số trang	18
Dung lượng	566,94 KB